Do powercords make a difference in sound?

08-15-11: Almarg
2)The differences reflect interactions between power cord characteristics, the characteristics of the particular incoming AC, and the particular components in the particular system, especially their power supplies.

3)There is no reason to expect much if any consistency of those interactions from system to system, and therefore no reason to expect much if any consistency or predictability of the sonic effects of a particular power cord.

4)While a power cord may have effects on the tonality of any given system, it is incorrect to attribute tonal characteristics to a power cord, because those tonal effects will vary from system to system.

It seems to me that, in the comments above, Al has identified one of the most recurring flaws in the thinking of many audiophiles, namely that…

If using component X results in audible difference Y, then audible difference Y is attributable to component X.

On the face of it, this statement appears to be not only true, but self evident. But under quite common circumstances, it is false. Specifically, it’s false when the audible difference resulting from the use of a component is an EXTRINSIC characteristic of that component.

An extrinsic characteristic is one that a thing has only IN RELATION TO other things. An intrinsic characteristic is one that a thing has INDEPENDENT OF other things. For example, mass is an intrinsic characteristic, weight is an extrinsic characteristic.

Judging from many posts I have seen on A’gon, there is a widespread tendency to believe that a component’s audible characteristic are INTRINSIC, even in cases where there is good reason to believe that they are EXTRINSIC. Al nicely summarized why the audible characteristics of power cords are largely extrinsic, but it seems to me that the same thing can be said of other types of components.

The tendency to regard a thing’s characteristics as intrinsic is a kind of “localization fallacy,” that is, a failure to recognize the characteristic’s dependence on the external things with which it interacts.

For the record, I’m guilty of the fallacy I’m describing, so I’m not casting stones. Just pointing out something that, I believe, results in needless confusion and disagreement here on A’gon. I suspect that the tendency to regard a thing’s characteristics as intrinsic is an inherent human bias. In any case, it seems to be an inherent audiophile bias.

FWIW.

Bryon

Wonderful comment and insight, Bryon. I could not be in more complete agreement.

Another example that provides a good illustration of what you are referring to would be digital cables. Certainly different digital cables can produce different sonic results in a given system. But those differences will be the result of a complex set of interactions between their characteristics, such as impedance accuracy, shield resistance, shield quality, propagation velocity, the particular length that is used, etc., and the characteristics of the components that the cable is connecting, such as impedance accuracy, signal risetimes and falltimes, ground loop susceptibility, jitter rejection capability, etc. Once again resulting in there being little likelihood of a given cable having similar sonic effects across different systems, a point that seems to be commonly under-recognized.

Thanks! Best regards,
-- Al

"Intrinsic" "extrinsic" "localization fallacy"

This PC business always gets heavy, dudes. ;)

Those are wonderful posts guys, but I'm not sure the argument is as impeccable in practice as on the page. I have to wonder if those positions are based more on theory or experience. In my own experience, although maybe some of this extrinsic/intrinsic business does apply (and I would agree that a good bit of it does apply when it comes to speaker cables in particular), I have pretty much found that any power cord's sonic signature migrates between components with the cord to a significant extent. Whether all of those marriages are happy, within a given system context, is of course a different question. (In my own system, I have actually settled on one preferred model of power cord for all applications, but I didn't always operate this way.)

If your take is based on experience, then my own feeling would be that what you're actually responding to is mostly a simple matter of how well (or not) any particular cord's sound complements any particular component's sound (and vice versa) -- rather than believing that, as Al puts it, "it is incorrect to attribute tonal characteristics to a power cord, because those tonal effects will vary from system to system". To me, this represents an extreme conclusion that, while it may be arrived at logically (if imperfectly), simply doesn't correspond to what I've observed.

I think that like anything else, this just boils down to questions of synergy and preference. (A listener's impression of a power cord might change with seating position!) But like yours, that's just my opinion...

Hi Zaikesman,

Let’s take some specific examples, relevant to power cords:

1)Power cords will differ somewhat in terms of resistance. That will result in differences in the ac voltages that appear at the power connectors of the components to which they are connected. The voltage differences will vary depending on the current draw of the components, as well as on the resistance of the cords. I’ll ignore the fact that those voItage differences may be overshadowed by fluctuations in the incoming line voltage that may occur from time to time, and from location to location. Is it expectable that the power cord-related voltage differences may be sonically significant? Certainly. Is it expectable that the resulting sonic differences among various components will be consistent and predictable?

I submit that experience and experimentation are irrelevant to the conclusion that the answer to that specific question is no. I say that as someone with multiple decades of experience designing advanced analog and digital circuits (not for audio). It would depend on how tightly regulated the power supplies of the components are, on the types of component parts that are used throughout the design, on the circuit topology that is used, on the thermal management provisions in the design, and on literally countless other component-dependent technical variables.

2)Power cords will differ somewhat in terms of shielding effectiveness. That will result in differences in the amount of rfi and noise that is presented to the power input of the connected component, as well as the amount of rfi and noise that is generated within that component that may couple from the power cord to other nearby components.

Is it expectable that those rfi and noise effects may have sonic consequences? Certainly. Is it expectable that those consequences will be consistent and predictable across different components, different systems, different physical arrangements of those components, and different interconnection configurations between those components? No way. Experience and experimentation are not necessary to reach that conclusion.

3)Power cords will differ in terms of inductance, which may, particularly in the case of power amplifiers, limit the speed with which abrupt changes in current demands can be responded to. I’ll ignore the fact that the Romex in the walls will probably impose limitations in that response that are much more significant. Given that, is it conceivable that those differences will have sonic consequences? Certainly. Is it expectable that those sonic consequences will be consistent among Class A designs, whose current draw variations are relatively small, Class D designs where those variations are very large, and Class AB designs that fall at various points in between? Not to mention that within each design class there are innumerable other relevant variables, such as energy storage provisions. Once again, I would assert that experience and experimentation are irrelevant to the conclusion that the answer is no.

I respect the sincerity of your conclusions, and your experience-based observations. I am one who tries to keep an open mind about such matters, BUT WITHIN FINITE LIMITS. I define those limits based on reasonableness, common sense, and my technical understanding of how things work.

I would add that performing a truly meaningful comparison of a multitude of power cords, encompassing a multitude of different components and systems, while carefully identifying and controlling extraneous variables, both technical ones and psychological ones, is a formidable task that would be impracticable for most audiophiles. I have certainly not attempted to do that. I recognize that there are some who would claim to have done that, and to have reached conclusions similar to yours. But with all due respect, I remain skeptical. (And I repeat, that skepticism is not about the existence of differences, but about their consistency and predictability from system to system, and about the degree of correlation between cost and performance).

If you or anyone else can cite additional technical factors beyond the three I mentioned above, that might account for why power cords may make differences that can be expected to be consistent and predictable across multiple component and system configurations, and that rise above the level of speculation or marketing pseudo-science, I would be happy to consider them and comment further.

Regards,
-- Al